Postgraduate Profiles

Monica Furaste Danilevicz

Monica Furaste Danilevicz profile photo

Thesis: Automatic detection model using deep learning algorithms for recognition of biotic and abiotic stress signs in canola leaf.

Agriculture has played a major role on the way societies have developed. As global population continues to grow, agriculture productivity must increase in order to meet new food demands. My PhD project will research the application of deep learning models for crop health monitoring, enabling the accurate identification of infectious diseases and abiotic stress (such as poor nutrition, water deficit and others). Initially the project will focus on canola leaf images, targeting the main visual indications of stress and disease infection. Deep learning algorithms have revolutionized data and image processing in several fields, due to their capability of autonomously identifing features that better represent each condition, generating a classification or prediction from observed data. A crop health monitoring tool will be based on a convolutional neural network and integrated to novel approaches, developing a new state-of-art tool for assessing plant health.

Why my research is important

Canola is Australia’s third largest broad acre crop, with Western Australia as the major producer. The development of a plant health monitoring application will support a rapid response to early stages of disease development. Pests and disease infections are widely recognized as one of the major constraints to food security, causing the loss of 20-40% of major crop yield worldwide. An automatic stress identification tool will enable farmers to treat the diseased area, avoiding its spread throughout the field. By enabling the targeted application of pesticides and fertilisers, the tool will decrease the environmental impacts associated with agriculture. It will also be useful for crop breeders to rapidly characterise plant responses to stress under field trials, assisting in the development of improved crop varieties that are better adapted to resist major diseases and changing climate conditions.

Funding

Apr 2019

Apr 2022